FlowCapX: Physics-Grounded Flow Capture with Long-Term Consistency

2025-10-27

AI Summary

Existing methods struggle to simultaneously reconstruct turbulent details and maintain physical consistency from sparse video inputs, limiting flow field reconstruction quality and downstream task performance. To address this, we propose a physics-guided multi-scale hybrid optimization framework: vorticity conservation constraints are imposed at coarse scales to ensure long-term physical fidelity, while fine-scale turbulent structures are preserved. Our approach integrates variational optimization, sparse observation alignment, and a vorticity-directed loss function, augmented by hierarchical supervision to improve training stability. The method significantly enhances both physical fidelity and geometric accuracy of velocity field reconstruction. It achieves state-of-the-art performance on standard benchmarks and enables high-fidelity flow analysis, enhanced tracer particle visualization, and reproducible flow simulation.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Search and Optimization: Learning to Search
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Security and Privacy: Large-scale security measurements

Abstract

We present FlowCapX, a physics-enhanced framework for flow reconstruction from sparse video inputs, addressing the challenge of jointly optimizing complex physical constraints and sparse observational data over long time horizons. Existing methods often struggle to capture turbulent motion while maintaining physical consistency, limiting reconstruction quality and downstream tasks. Focusing on velocity inference, our approach introduces a hybrid framework that strategically separates representation and supervision across spatial scales. At the coarse level, we resolve sparse-view ambiguities via a novel optimization strategy that aligns long-term observation with physics-grounded velocity fields. By emphasizing vorticity-based physical constraints, our method enhances physical fidelity and improves optimization stability. At the fine level, we prioritize observational fidelity to preserve critical turbulent structures. Extensive experiments demonstrate state-of-the-art velocity reconstruction, enabling velocity-aware downstream tasks, e.g., accurate flow analysis, scene augmentation with tracer visualization and re-simulation.

Problem

• Reconstructing flow from sparse video with physical constraints
• Maintaining long-term consistency in turbulent motion capture
• Resolving sparse-view ambiguities while preserving turbulent structures

Innovation

• Hybrid framework separates representation across spatial scales
• Optimization aligns long-term observation with physics-grounded velocity
• Emphasizes vorticity-based constraints for physical fidelity